https://github.com/mightypixel/thanksmate
The project aims to provide a simple way to give appreciation of others who helped you in some way.
https://github.com/mightypixel/thanksmate
Last synced: 9 months ago
JSON representation
The project aims to provide a simple way to give appreciation of others who helped you in some way.
- Host: GitHub
- URL: https://github.com/mightypixel/thanksmate
- Owner: MightyPixel
- License: agpl-3.0
- Created: 2015-04-24T17:47:56.000Z (about 11 years ago)
- Default Branch: master
- Last Pushed: 2016-02-17T09:04:03.000Z (over 10 years ago)
- Last Synced: 2025-03-17T12:47:22.630Z (over 1 year ago)
- Language: JavaScript
- Homepage:
- Size: 2.47 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# About the Project
## Abstract
The project aims to provide a simple way to give appreciation of others who helped you in some way.
## Flow
Users register with a photo. Every user have the ability to say "Thanks" to someone, even if he isn't registered as an user. Saying thanks consists of a) comment what happened b) a photo of the user who helped you c) tags (optional). The photo is used to find the user who is responsible for the good deed and to give him a reward. If no user with such photo is found we create an anonymous user with that photo, once the real person registers the photo from the registration is used to match him to the anonymous user.
## Face Recognition
__Yet to be solved.__
The application requires face recognition with only one sample [(single-sample problem)](http://yima.csl.illinois.edu/psfile/cvpr13_robust_alignment.pdf)
There are some simple techniques such as applying grayscale filter, cropping and aligning the face, extracting features that yield ~75% match rate - not sufficient for production use.
Alternative approach would be Facebook's
[DeepFace](https://research.facebook.com/publications/deepface-closing-the-gap-to-human-level-performance-in-face-verification/)
## Reward Algorithm
The rewards are given based on the actions (Thanks given). The value of the reward is calculated with Katz's graph algorithm. This approach is used in social networks and page rank algorithms.
## Used Technologies
- NodeJS
- SailsJS
- Mongo
- OpenCV
- vis.js
### Setup
install NodeJS
install mongo
install OpenCV
run npm install
run sails lift
visit http://localhost:1337/